At present, platelet transfusions are limited by their short storage half-life, variable quality and quantity, high- risk of bacterial contamination, and frequency of patients developing inhibitors to the transfused platelets. We propose a strategy for generating from human embryonic stem cells (hESC) ex vivo, non-donor-dependent, sources of platelets. These studies would also allow the development of a novel strategy for targeted delivery of therapeutics to sites of vascular injury. The specific aims in developing the above strategy are as follows: Specific Aim 1. Develop large-scale production methods for platelets from hESC. Our group has shown that murine ES cells deficient in the hematopoietic transcription factor GATA-1, when grown in the presence of thrombopoietin (TPO) and OP9 cells become OP9 independent/TPO-dependent and appear to have properties of a megakaryocyte-erythrocyte progenitor (MEP) cell. These MEP-like cells were termed G1ME cells and upon re-expression of GATA-1, ~60% develop into identifiable megakaryocytes. We propose to establish a similar cell line from hESC. Alternatively, beginning with hESC, we will optimize megakaryocyte cell production directly from hESC. Specific Aim 2. Develop strategies for obtaining significant in vivo increases in platelet counts. To date, where megakaryocytes release platelets in vivo has not been clearly demonstrated. Megakaryocytes may release platelets within the marrow or after migration to the lungs. We will pursue approaches for efficient ex vivo formation of functional platelets for infusion. We have pursued an alternative model for platelet formation by infusing mature murine megakaryocytes demonstrating vigorous, almost- immediate platelet release in recipient mice. In this application, we will pursue both approaches for delivery of transfused platelets: 1) Generating ex vivo platelets;and 2) infusing megakaryocytes. Efficacy studies will involve infusion of these platelets/megakaryocytes in a recently described NOD/SCID animal model for human platelet infusion studies and examining the qualitative and quantitative response. 3) Specific Aim 3. Develop modified platelets for targeted hemostatic, fibrinolytic and angiogenic purposes. We have published proof-of- principle studies in mice that platelet-delivery of an ectopic protein can be useful in several clinical settings for targeted delivery. We now propose to use the glycoprotein (GP) Ib1 promoter-driven construct to drive high levels of tissue-specific expression of uPA within developing megakaryocytes and to transfuse modified platelets/megakaryocytes for targeted delivery of this fibrinolytic agent using a modification of the above NOD/SCID mouse platelet transfusion model for efficacy of the uPA-containing platelets to lyse thrombi in a targeted fashion. We not only believe that the studies in this application will be of great clinical utility, but be easily incorporated into consortiums involving characterization of hESC, generating self-replicating progenitor lineages, characterizing terminally differentiated cells from hESC and modifying the nature of terminally differentiated cells.